Luciferases are enzymes that catalyze the oxidation of a substrate (e.g., luciferin) with the concomitant release of photons of light. Luciferases have been isolated from numerous species, including Coleopteran arthropods and many sea creatures. Because it is easily detectable and its activity can be quantified with high precision, luciferases have been used widely to study gene expression and protein localization. Unlike green fluorescent protein (GFP), which requires up to 30 minutes to form its chromophore, the products of luciferases can be detected immediately upon completion of synthesis of the polypeptide chain (if substrate and oxygen are also present). In addition, no post-translational modifications are required for enzymatic activity, and the enzyme contains no prosthetic groups, bound cofactors, or disulfide bonds. Luciferases are useful reporters in numerous species and in a wide variety of cells.
Luciferases possess additional features that render them particularly useful as reporter molecules for biosensing, i.e., molecules which reveal molecular properties of a system. Biosensors (i.e., sensors which comprise a biological component) generally function by means of a two-step process: signal generation mediated through a biological component, and signal transduction and/or amplification through an electrical component. Signal generation is typically achieved through binding, energy transfer or catalysis. Signal generation by enzymatic catalysis can be particularly useful due to the inherent efficiency and specificity of these chemical processes. Most catalytic reactions generate less than the energy of hydrolysis for two molecules of ATP, or about 70 kJ/mole. However, the luminescence elicited by luciferases has much higher energy content. For instance, the reaction catalyzed by firefly luciferase (560 nm) emits 214 kJ/mole of energy. Furthermore, luciferases are also highly efficient at converting chemical energy into photons, i.e., they have high quantum yields. Luciferases are thus extremely efficient for generating detectable signals.
Luciferase biosensors have been described. For example, Sala-Newby et al. (Biochem J. 279:727 (1991)) disclose that a Photinus pyralis luciferase cDNA was modified to generate cyclic AMP-dependent protein kinase phosphorylation sites. In particular, a valine at position 217 was mutated to arginine to generate a site, RRFS (SEQ ID NO:1), and the heptapeptide kemptide, the phosphorylation site of the porcine pyruvate kinase, was added at the N- or C-terminus of the luciferase. Sala-Newby et al. relate that the proteins carrying phosphorylation sites were characterized for their specific activity, pI, effect of pH on the color of the light emitted, and effect of the catalytic subunit of protein kinase A in the presence of ATP. They found that only one of the recombinant proteins (RRFS; SEQ ID NO:1) was significantly different from wild-type luciferase and that the RRFS (SEQ ID NO:1) mutant had a lower specific activity, lower pH optimum, emitted greener light at low pH and, when phosphorylated, decreased its activity by up to 80%. It is disclosed that the latter effect was reversed by phosphatase.
Waud et al. (BBA 1292: 89 (1996)) engineered protein kinase recognition sequences and proteinase sites into a Photinus pyralis luciferase cDNA. Two domains of the luciferase were modified by Waud et al.; one between amino acids 209 and 227 and the other at the C-terminus between amino acids 537 and 550. Waud et al. disclose that the mutation of amino acids between residues 209 and 227 reduced bioluminescent activity to less than 1% of wild-type recombinant, while engineering peptide sequences at the C-terminus resulted in specific activities ranging from 0.06%-120% of the wild-type recombinant luciferase. Waud et al. also disclose that addition of a cyclic AMP dependent protein kinase catalytic subunit to a variant luciferase incorporating the kinase recognition sequence, LRRASLG (SEQ ID NO:2), with a serine at amino acid position 543, resulted in a 30% reduction activity. Alkaline phosphatase treatment restored activity. Waud et al. further disclose that the bioluminescent activity of a variant luciferase containing a thrombin recognition sequence, LVPRES (SEQ ID NO:3), with the cleavage site positioned between amino acids 542 and 543, decreased by 50% when incubated in the presence of thrombin.
Ozawa et al. (Analytical Chemistry 73: 2516 (2001)) describe a biosensor based on protein splicing-induced complementation of rationally designed fragments of firefly luciferase. Protein splicing is a posttranslational protein modification through which inteins (internal proteins) are excised out from a precursor fusion protein, ligating the flanking exteins (external proteins) into a contiguous polypeptide. It is disclosed that the N- and C-terminal intein DnaE from Synechocystis sp. PCC6803 were each fused respectively to N- and C-terminal fragments of a luciferase. Protein-protein interactions trigger the folding of DnaE intein, resulting in protein splicing, and thereby the extein of ligated luciferase recovers its enzymatic activity. Ozawa et al. disclose that the interaction between known binding partners, phosphorylated insulin receptor substrate 1 (IRS-1) and its target N-terminal SH2 domain of PI 3-kinase, was monitored using a split luciferase in the presence insulin.
Paulmurugan et al. (PNAS USA 99: 3105 (2002)) employed a split firefly luciferase-based assay to monitor the interaction of two proteins, i.e., MyoD and Id, in cell cultures and in mice using both complementation strategy and an intein-mediated reconstitution strategy. To retain reporter activity, in the complementation strategy, fusion proteins need protein interaction, i.e., via the interaction of the protein partners MyoD and Id, while in the reconstitution strategy, the new complete beetle luciferase formed via intein-mediated splicing maintains it activity even in the absence of a continuing interaction between the protein partners.
A protein fragment complementation assay is disclosed in Michnick et al. (U.S. Pat. Nos. 6,270,964, 6,294,330 and 6,428,951). Specifically, Michnick describe a split murine dihydrofolate reductase (DHFR) gene-based assay in which an N-terminal fragment of DHFR and a C-terminal fragment of DHFR are each fused to a GCN4 leucine zipper sequence. DHFR activity was detected in cells which expressed both fusion proteins. Michnick et al. also describe another complementation approach in which nested sets of S1 nuclease generated deletions in the aminoglycoside kinase (AK) gene are introduced into a leucine zipper construct, and the resulting sets of constructs introduced to cells and screened for AK activity.